Support capturing groups in Parameter regex (#204)

- fix `{string}` parameter returning enclosing quotes (cucumber-rs/cucumber-expressions#7)

Author: ilslv

CHANGELOG.md

@@ -6,17 +6,27 @@ All user visible changes to `cucumber` crate will be documented in this file. Th
 
 
 
-## [0.11.4] · 2022-02-??
-[0.11.4]: /../../tree/v0.11.4
+## [0.12.0] · 2022-02-??
+[0.12.0]: /../../tree/v0.12.0
 
-[Diff](/../../compare/v0.11.3...v0.11.4) | [Milestone](/../../milestone/9)
+[Diff](/../../compare/v0.11.3...v0.12.0) | [Milestone](/../../milestone/9)
+
+### BC Breaks
+
+- `step::Context::matches` now has regex group name in addition to captured value. ([#204])
+
+### Added
+
+- Support for capturing groups in `Parameter` regex. ([#204], [cucumber-rs/cucumber-expressions#7])
 
 ### Fixed
 
 - Book examples failing on Windows. ([#202], [#200])
 
 [#200]: /../../issues/200
 [#202]: /../../pull/202
+[#204]: /../../pull/204
+[cucumber-rs/cucumber-expressions#7]: https://github.com/cucumber-rs/cucumber-expressions/issues/7
 
 
 

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cucumber"
-version = "0.11.3"
+version = "0.12.0-dev"
 edition = "2021"
 rust-version = "1.57"
 description = """\
@@ -53,8 +53,8 @@ regex = "1.5"
 sealed = "0.4"
 
 # "macros" feature dependencies.
-cucumber-codegen = { version = "0.11", path = "./codegen", optional = true }
-cucumber-expressions = { version = "0.1", features = ["into-regex"], optional = true }
+cucumber-codegen = { version = "0.12.0-dev", path = "./codegen", optional = true }
+cucumber-expressions = { version = "0.2", features = ["into-regex"], optional = true }
 inventory = { version = "0.2", optional = true }
 
 # "output-json" feature dependencies.

book/src/rec/writing_capturing_multiple_groups.gif

@@ -249,7 +249,7 @@ fn feed_cat(world: &mut AnimalWorld, times: u8) {
 
 ### Custom [parameters]
 
-Another useful advantage of using [Cucumber Expressions][expr] is an ability to declare and reuse  [custom parameters] in addition to [default ones][parameters].
+Another useful advantage of using [Cucumber Expressions][expr] is an ability to declare and reuse [custom parameters] in addition to [default ones][parameters].
 
 ```rust
 # use std::{convert::Infallible, str::FromStr};
@@ -335,6 +335,96 @@ fn hungry_cat(world: &mut AnimalWorld, state: State) {
 
 ![record](../rec/writing_capturing_both.gif)
 
+> __TIP__: In case [regex] of a [custom parameter][custom parameters] consists of several capturing groups, only the first non-empty match will be returned. 
+
+```rust
+# use std::{convert::Infallible, str::FromStr};
+#
+# use async_trait::async_trait;
+# use cucumber::{given, then, when, World, WorldInit};
+use cucumber::Parameter;
+
+# #[derive(Debug)]
+# struct Cat {
+#     pub hungry: Hungriness,
+# }
+#
+# impl Cat {
+#     fn feed(&mut self) {
+#         self.hungry = Hungriness::Satiated;
+#     }
+# }
+#
+#[derive(Debug, Eq, Parameter, PartialEq)]
+#[param(regex = "(hungry)|(satiated)|'([^']*)'")]
+// We want to capture without quotes  ^^^^^^^
+enum Hungriness {
+    Hungry,
+    Satiated,
+    Other(String),
+}
+
+// NOTE: `Parameter` requires `FromStr` being implemented.
+impl FromStr for Hungriness {
+    type Err = String;
+
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        Ok(match s {
+            "hungry" => Self::Hungry,
+            "satiated" => Self::Satiated,
+            other => Self::Other(other.to_owned()),
+        })
+    }
+}
+#
+# #[derive(Debug, WorldInit)]
+# pub struct AnimalWorld {
+#     cat: Cat,
+# }
+#
+# #[async_trait(?Send)]
+# impl World for AnimalWorld {
+#     type Error = Infallible;
+#
+#     async fn new() -> Result<Self, Infallible> {
+#         Ok(Self {
+#             cat: Cat {
+#                 hungry: Hungriness::Satiated,
+#             },
+#         })
+#     }
+# }
+
+#[given(expr = "a {hungriness} cat")]
+fn hungry_cat(world: &mut AnimalWorld, hungry: Hungriness) {
+    world.cat.hungry = hungry;
+}
+
+#[then(expr = "the cat is {string}")]
+fn cat_is(world: &mut AnimalWorld, other: String) {
+    assert_eq!(world.cat.hungry, Hungriness::Other(other));
+}
+#
+# #[when(expr = "I feed the cat {int} time(s)")]
+# fn feed_cat(world: &mut AnimalWorld, times: u8) {
+#     for _ in 0..times {
+#         world.cat.feed();
+#     }
+# }
+# 
+# #[then("the cat is not hungry")]
+# fn cat_is_fed(world: &mut AnimalWorld) {
+#     assert_eq!(world.cat.hungry, Hungriness::Satiated);
+# }
+#
+# #[tokio::main]
+# async fn main() {
+#     AnimalWorld::run("tests/features/book/writing/capturing_multiple_groups.feature").await;
+# }
+```
+
+![record](../rec/writing_capturing_multiple_groups.gif)
+
 
 
 

book/src/writing/capturing.md

@@ -12,7 +12,7 @@ publish = false
 [dependencies]
 async-trait = "0.1"
 clap = { version = "3.0", features = ["derive"] }
-cucumber = { version = "0.11", path = "../..", features = ["output-json", "output-junit"] }
+cucumber = { version = "0.12.0-dev", path = "../..", features = ["output-json", "output-junit"] }
 futures = "0.3"
 humantime = "2.1"
 once_cell = { version = "1.8", features = ["parking_lot"] }

book/tests/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cucumber-codegen"
-version = "0.11.3" # should be the same as main crate version
+version = "0.12.0-dev" # should be the same as main crate version
 edition = "2021"
 rust-version = "1.57"
 description = "Code generation for `cucumber` crate."
@@ -21,7 +21,7 @@ exclude = ["/tests/"]
 proc-macro = true
 
 [dependencies]
-cucumber-expressions = { version = "0.1", features = ["into-regex"] }
+cucumber-expressions = { version = "0.2", features = ["into-regex"] }
 inflections = "1.1"
 itertools = "0.10"
 proc-macro2 = "1.0.28"

codegen/Cargo.toml

@@ -213,18 +213,62 @@ impl Step {
         if is_regex_or_expr {
             if let Some(elem_ty) = find_first_slice(&func.sig) {
                 let addon_parsing = Some(quote! {
-                    let __cucumber_matches = __cucumber_ctx
+                    let mut __cucumber_matches = ::std::vec::Vec::with_capacity(
+                        __cucumber_ctx.matches.len().saturating_sub(1),
+                    );
+                    let mut __cucumber_iter = __cucumber_ctx
                         .matches
                         .iter()
                         .skip(1)
-                        .enumerate()
-                        .map(|(i, s)| {
+                        .enumerate();
+                    while let Some((i, (cap_name, s))) =
+                        __cucumber_iter.next()
+                    {
+                        // Special handling of `cucumber-expressions`
+                        // `parameter` with multiple capturing groups.
+                        let prefix = cap_name
+                            .as_ref()
+                            .filter(|n| n.starts_with("__"))
+                            .map(|n| {
+                                let num_len = n
+                                    .chars()
+                                    .skip(2)
+                                    .take_while(|&c| c != '_')
+                                    .map(char::len_utf8)
+                                    .sum::<usize>();
+                                let len = num_len + b"__".len();
+                                n.split_at(len).0
+                            });
+
+                        let to_take = __cucumber_iter
+                            .clone()
+                            .take_while(|(_, (n, _))| {
+                                prefix
+                                    .zip(n.as_ref())
+                                    .filter(|(prefix, n)| n.starts_with(prefix))
+                                    .is_some()
+                            })
+                            .count();
+
+                        let s = ::std::iter::once(s.as_str())
+                            .chain(
+                                __cucumber_iter
+                                    .by_ref()
+                                    .take(to_take)
+                                    .map(|(_, (_, s))| s.as_str()),
+                            )
+                            .fold(None, |acc, s| {
+                                acc.or_else(|| (!s.is_empty()).then(|| s))
+                            })
+                            .unwrap_or_default();
+
+                        __cucumber_matches.push(
                             s.parse::<#elem_ty>().unwrap_or_else(|e| panic!(
                                 "Failed to parse element at {} '{}': {}",
                                 i, s, e,
                             ))
-                        })
-                        .collect::<Vec<_>>();
+                        );
+                    }
                 });
                 let func_args = func
                     .sig
@@ -319,11 +363,49 @@ impl Step {
             );
 
             quote! {
-                let #ident = __cucumber_iter
-                    .next()
-                    .expect(#not_found_err)
-                    .parse::<#ty>()
-                    .expect(#parsing_err);
+                let #ident = {
+                    let (cap_name, s) = __cucumber_iter
+                        .next()
+                        .expect(#not_found_err);
+                    // Special handling of `cucumber-expressions` `parameter`
+                    // with multiple capturing groups.
+                    let prefix = cap_name
+                        .as_ref()
+                        .filter(|n| n.starts_with("__"))
+                        .map(|n| {
+                            let num_len = n
+                                .chars()
+                                .skip(2)
+                                .take_while(|&c| c != '_')
+                                .map(char::len_utf8)
+                                .sum::<usize>();
+                            let len = num_len + b"__".len();
+                            n.split_at(len).0
+                        });
+
+                    let to_take = __cucumber_iter
+                        .clone()
+                        .take_while(|(n, _)| {
+                            prefix.zip(n.as_ref())
+                                .filter(|(prefix, n)| n.starts_with(prefix))
+                                .is_some()
+                        })
+                        .count();
+
+                    ::std::iter::once(s.as_str())
+                        .chain(
+                            __cucumber_iter
+                                .by_ref()
+                                .take(to_take)
+                                .map(|(_, s)| s.as_str()),
+                        )
+                        .fold(
+                            None,
+                            |acc, s| acc.or_else(|| (!s.is_empty()).then(|| s)),
+                        )
+                        .unwrap_or_default()
+                };
+                let #ident = #ident.parse::<#ty>().expect(#parsing_err);
             }
         };
 
@@ -533,7 +615,7 @@ impl<'p> Parameters<'p> {
         self.0
             .iter()
             .map(|par| {
-                let name = par.param.0.fragment();
+                let name = par.param.input.fragment();
                 let ty = &par.ty;
 
                 if DEFAULT_PARAMETERS.contains(name) {
@@ -619,7 +701,7 @@ impl<'p> Parameters<'p> {
             .0
             .iter()
             .filter_map(|par| {
-                let name = par.param.0.fragment();
+                let name = par.param.input.fragment();
                 (!DEFAULT_PARAMETERS.contains(name)).then(|| (*name, &par.ty))
             })
             .unzip();

codegen/src/attribute.rs

@@ -309,7 +309,9 @@ steps!(given, when, then);
 ///
 /// - `#[param(regex = "regex")]`
 ///
-///   [`Regex`] to match this parameter. Shouldn't contain any capturing groups.
+///   [`Regex`] to match this parameter. Usually shouldn't contain any capturing
+///   groups, but in case it requires to do so, only the first non-empty group
+///   will be matched as the result.
 ///
 /// - `#[param(name = "name")]` (optional)
 ///

codegen/src/lib.rs

@@ -68,12 +68,6 @@ impl TryFrom<syn::DeriveInput> for Definition {
             // TODO: Use "{e}" syntax once MSRV bumps above 1.58.
             syn::Error::new(attrs.regex.span(), format!("Invalid regex: {}", e))
         })?;
-        if regex.captures_len() > 1 {
-            return Err(syn::Error::new(
-                attrs.regex.span(),
-                "Regex shouldn't contain any capturing groups",
-            ));
-        }
 
         let name = attrs.name.as_ref().map_or_else(
             || to_lower_case(&input.ident.to_string()),
@@ -156,6 +150,27 @@ mod spec {
         );
     }
 
+    #[test]
+    fn derives_impl_with_capturing_group() {
+        let input = parse_quote! {
+            #[param(regex = "(cat)|(dog)")]
+            struct Animal;
+        };
+
+        let output = quote! {
+            #[automatically_derived]
+            impl ::cucumber::Parameter for Animal {
+                const REGEX: &'static str = "(cat)|(dog)";
+                const NAME: &'static str = "animal";
+            }
+        };
+
+        assert_eq!(
+            super::derive(input).unwrap().to_string(),
+            output.to_string(),
+        );
+    }
+
     #[test]
     fn derives_impl_with_generics() {
         let input = parse_quote! {
@@ -215,17 +230,21 @@ mod spec {
     }
 
     #[test]
-    fn errors_on_capture_groups_in_regex() {
+    fn invalid_regex() {
         let input = parse_quote! {
-            #[param(regex = "(cat|dog)")]
+            #[param(regex = "(cat|dog")]
             struct Parameter;
         };
 
         let err = super::derive(input).unwrap_err();
 
         assert_eq!(
             err.to_string(),
-            "Regex shouldn't contain any capturing groups",
+            "\
+Invalid regex: regex parse error:
+    (cat|dog
+    ^
+error: unclosed group",
         );
     }
 }